Answers are grounded in this paper and cite their sources.
Retinal optical coherence tomography imaging is an important diagnostic tool for early detection and monitoring of various retinal disorders. This study aims to improve diagnostic accuracy by leveraging deep learning methods such as VGG16 and InceptionV3 models to classify retinal OCT images automatically, ultimately enhancing ocular diagnostics and clinical decision-making. Utilizing a dataset comprising approximately 85,000 high-resolution images across primary retinal condition categories including diabetic macular edema (DME), age-related macular degeneration (AMD), Choroidal Neovascularization (CNV), and healthy retinas, the efficacy of these deep learning models is assessed in this comprehensive study. Results demonstrate that VGG16 excels in CNV classification and InceptionV3 is proficient in identifying normal retinal conditions. Comparative analysis underscores their distinct strengths, with both models achieving an impressive overall accuracy of 92.76%. These findings demonstrate the potential of deep learning for automating OCT image interpretation and enhancing ocular diagnostics, thereby improving patient outcomes and clinical decision-making resulting in advanced healthcare practices.
No takes yet. Share an insight, caveat, or question.
Rithani et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: